Dr Sushil Kumar

Scopus Publications

Optimizing Air Pollution Prediction With Random Forest Algorithm
Sukhendra Singh, Manoj Kumar, Birendra Kumar Verma, Sushil Kumar
Aerosol Science and Engineering, 2026
Advances in Deep Learning Techniques for Breast Cancer Classification: A Comprehensive Review
Kanika Kansal, Sushil Kumar, Kajal Kansal
Archives of Computational Methods in Engineering, 2026
Advanced gesture recognition in Indian sign language using a synergistic combination of YOLOv10 with Swin Transformer model
Umang Rastogi, Rajendra Prasad Mahapatra, Sushil Kumar
Scientific Reports, 2025
Communication between deaf or mute individuals and hearing persons is often hindered by the lack of mutual understanding of sign or vocal language. To bridge this gap, Indian Sign Language Recognition (ISLR) systems are essential. This paper proposes a real-time ISLR framework based on the YOLOv10-ST model, which integrates the Swin Transformer into the YOLOv10 architecture for enhanced feature extraction. The model also incorporates Mish activation to improve gradient flow and detection accuracy. A custom dataset comprising 15, 000 static images (1, 000 per sign for 15 signs) and 35 dynamic videos (covering 7 sign classes) was used for training and evaluation. Experimental results demonstrate high performance, with the model achieving 97.50% precision, 98.10% recall, and 96.58% F1-score for image-based sign recognition, and 95.24% precision, 96.00% recall, and 95.87% F1-score for video-based gestures. The model also achieves a mean Average Precision (mAP) of 97.62% and real-time inference speeds of 48.7 FPS. Ablation studies validate the contributions of Swin Transformer and Mish activation, while paired t-tests confirm statistical significance (p $$< 0.005$$ ). The experimental findings demonstrate that the YOLOv10-ST model efficiently recognizes static and dynamic ISL in real time with minimal computational overhead.
Advancements in Machine Learning Techniques for Hand Gesture-Based Sign Language Recognition: A Comprehensive Review
Umang Rastogi, Rajendra Prasad Mahapatra, Sushil Kumar
Archives of Computational Methods in Engineering, 2025
A Comprehensive Analysis of Quaternion Deep Neural Networks: Architectures, Applications, Challenges, and Future Scope
Sukhendra Singh, Sushil Kumar, B. K. Tripathi
Archives of Computational Methods in Engineering, 2025
Residual quadratic encoder–decoder architecture for semantic segmentation of satellite images
Neha Bagwari, Vivek Singh Verma, Sushil Kumar
Signal Image and Video Processing, 2025
Air Pollution Forecasting Using Genetic Algorithm and LSTM Deep Learning Technique
Umang Rastogi, Sushil Kumar, Satya Prakash Maurya, Akhilesh Kumar Yadav
Aerosol Science and Engineering, 2025
Progressing Alzheimer's Diagnosis with Ensemble CNN Model
Vivek Yadav, Vaibhav, Rohit, Sushil Kumar, Umang Rastogi
Icdt 2025 3rd International Conference on Disruptive Technologies, 2025
This paper proposes an Alzheimer's disease prediction model using CNN, implemented on a system with AMD Radeon Vega 8 Graphics to improve computational traceability. The research is encouraging because the implicates of the model touched the 97% accuracy of the disease prediction by MRI images of patients. The integration of the Vega 8 GPU allows the system to process large datasets and perform computations at a significantly faster rate, thereby reducing training time. The MRI dataset used in this study is highly imbalanced, with four classes: Non-Demented or Non-D, Very Mild Dementia or V.M.D, Mild Dementia or M.D, Moderate Demented or M.D.A. CNN was chosen for its capability in highlight extraction and learning from MRI pictures, without requiring human interference. The show was run with parallel and multiclass datasets, and IT got 97% exactness. The systematic analysis of the test shows that the system can diagnose Alzheimer's right at the initial stage of the disease development. This approach shows how CNNs with the aid of AMD Radeon Vega 8 Graphics perform well in automating Alzheimer's diagnosis while breaking the barriers associated with manual feature extraction and orthodox dependence on experts.
Book Recommendation System Using Hybrid Content and Collaborative Filtering Techniques
Urvi Gupta, Tripti Singh, Vidyush Singh, Umang Rastogi, Sushil Kumar
Icdt 2025 3rd International Conference on Disruptive Technologies, 2025
In this study, the process of producing a book recommendation system outlines the process of finding books to provide personalized suggestions according to user preference and behavior. The system increases user satisfaction by tailoring recommendations to their tastes, improving the overall reading experience, and enhancing engagement and retention on plat-forms. It tracks users with collaborative filtering. interactions, content-based filtering to study book characteristics such as genre and author, and mixed methods to combine both techniques that will help ensure accuracy and circumvent similar limitations of cold start problems. Machine learning models-, pre-processing of data algorithms, and accuracy and recall measures of analysis ensure the effectiveness of the system in providing individualized and relevant recommendations that are 85% accurate. The findings reveal test the system's capacity to generate good suggestions, improving user experience and engagement. Improvements in the future could include real-time recommendations and broader feedback integration towards further optimization in accuracy and usability.
A Hybrid Approach based on Haar Cascade, Softmax, and CNN for Human Face Recognition
Pancham Singh, Mrignainy Kansal, Rajeev Kumar Singh, Sushil Kumar, Chelsi Sen
Journal of Scientific and Industrial Research, 2024
Face recognition has been studied long but it is still an important and current research field in deep learning, computer vision, and forensics. There are several applications such as group action systems, human-machine interaction, and security systems, where face recognition is of vital importance. It is noticed that the algorithms based on Deep Learning (DL) have shown higher performances, stipulation of accuracy, and processing speed as compared to traditional machine learning algorithms. With its dominant methodology in deep learning, the Convolutional Neural Network (CNN) has contributed immensely to face recognition. In this paper, a novel hybrid version of the deep learning algorithm containing Haar Cascade, SoftMax, and CNN components is proposed. It provides promising results for applications based on the recognition of human faces. In the experiments, the accuracy of this hybrid algorithm is achieved at 99.95%, which is significantly higher than existing Viola-Jonas and Principal Component Analysis (PCA), which have accuracy rates of 74.38% and 81.81% respectively. However, the accuracy of our proposed algorithm close to Linear Discriminant Analysis (LDA) at 95.45%, and SoftMax and CNN at 94%. In this paper, the proposed hybrid deep learning algorithm improves the result performance and is compared with some existing techniques for face recognition.
Modeling and Early Diagnosis of Alzheimer’s Disease Using Recurrent Neural Network
Mitu Ranjan, Sushil Kumar
Lecture Notes in Networks and Systems, 2024
Modeling of progressive Alzheimer’s disease using machine learning algorithms
Mitu Ranjan, Sushil Kumar
Artificial Intelligence Blockchain Computing and Security Proceedings of the International Conference on Artificial Intelligence Blockchain Computing and Security Icabcs 2023, 2024
Decentralized System for Counterfeit Product Detection using Blockchain Technology
Umang Rastogi, Dhruv Jain, Dhruv Gaur, Sushil Kumar
2024 International Conference on Intelligent Systems for Cybersecurity Iscs 2024, 2024
Comparative Analysis of U-Net Models Using ResNet34, InceptionV3, and VGG16 for the Processing of Satellite Images
Sushil Kumar, Anubhav Yadav, Ajay Varshney, Aayush Sharma, Shivani Shivani
2024 IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation Iatmsi 2024, 2024
A Comprehensive Approach to Stress Detection and Management Using Machine Learning, Deep Learning, and Chatbot Integration
Amisha Singh, Akshita Agarwal, Ajeet Kumar, Umang Rastogi, Sushil Kumar
2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
Human Detection to Mitigate Excessive Electrical Energy Usage in Communal Spaces
Aabhas Bisaria, Devesh Chauhan, Harsh Varshney, Sushil Kumar, Umang Rastogi, Vineet Sharma
2024 International Conference on Intelligent Systems for Cybersecurity Iscs 2024, 2024
Enhancing Trust and Transparency with Blockchain-Powered Remote Electronic Voting
Sushil Kumar, Shubham Verma, Shaurya Gupta, Shivam Shrivas, Vimal Kumar Dubey, Umang Rastogi
2024 International Conference on Control Computing Communication and Materials Iccccm 2024, 2024
A Comprehensive Review on Segmentation Techniques for Satellite Images
Neha Bagwari, Sushil Kumar, Vivek Singh Verma
Archives of Computational Methods in Engineering, 2023
A Comprehensive Review on the Advancement of High-Dimensional Neural Networks in Quaternionic Domain with Relevant Applications
Sushil Kumar, Umang Rastogi
Archives of Computational Methods in Engineering, 2023
A novel non-linear neuron model based on multiplicative aggregation in quaternionic domain
Sushil Kumar, Rishitosh Kumar Singh, Aryan Chaudhary
Complex and Intelligent Systems, 2023
Modeling of progressive Alzheimer’s disease using machine learning algorithms
Mitu Ranjan, Sushil Kumar
Artificial Intelligence Blockchain Computing and Security Volume 1, 2023
On the learning machine in quaternionic domain and its application
Sushil Kumar, Bipin Kumar Tripathi
International Journal of Advanced Intelligence Paradigms, 2023
Utilizing QUALNET for Analyzing MANET Protocols
Gaurav Singh Rawat, Sushil Kumar, Umang Rastogi
2022 3rd International Conference on Issues and Challenges in Intelligent Computing Techniques Icict 2022, 2022
Comparative Analysis of Differential Evolution Algorithm Using Shannon, Fuzzy, and Cosine Similarity Entropy Functions for Satellite Image Segmentation
Neha Bagwari, Sushil Kumar, Vivek Singh Verma
2022 3rd International Conference on Issues and Challenges in Intelligent Computing Techniques Icict 2022, 2022
Feature Extraction in Arabic Sign Language using Hand and Wrist Localization Techniques
Umang Rastogi, Sushil Kumar, Gaurav Rawat
Pdgc 2022 2022 7th International Conference on Parallel Distributed and Grid Computing, 2022
On the Learning Machine with Amplificatory Neuron in Complex Domain
Sushil Kumar, Rishitosh Kumar Singh, Aryan Chaudhary
Arabian Journal for Science and Engineering, 2020
Root-Power Mean Aggregation-Based Neuron in Quaternionic Domain
Sushil Kumar, Bipin Kumar Tripathi
IETE Journal of Research, 2019
On the learning machine with quaternionic domain neural network and its high-dimensional applications
Sushil Kumar, Bipin Kumar Tripathi
Journal of Intelligent and Fuzzy Systems, 2019
On the learning machine with compensatory aggregation based neurons in quaternionic domain
Sushil Kumar, Bipin Kumar Tripathi
Journal of Computational Design and Engineering, 2019
Medical data clustering based on particle swarm optimisation and genetic algorithm
Indresh Kumar Gupta, Vikash Yadav, Sushil Kumar
International Journal of Advanced Intelligence Paradigms, 2019
High-dimensional information processing through resilient propagation in quaternionic domain
Sushil Kumar, Bipin Kumar Tripathi
Journal of Industrial Information Integration, 2018
On the root-power mean aggregation based neuron in quaternionic domain
Harcourt Butler Technical University/Department of Computer Science & Engineering, Kanpur, 208002, India, Sushil Kumar, Bipin K. Tripathi
International Journal of Intelligent Systems and Applications, 2018
Machine learning with resilient propagation in quaternionic domain
Harcourt Butler Technical University, Sushil Kumar, Bipin Tripathi, Harcourt Butler Technical University
International Journal of Intelligent Engineering and Systems, 2017

Dr Sushil Kumar

RESEARCH, TEACHING, or OTHER INTERESTS

Scopus Publications